Car Detection Real Time With YoloV5 and Interface Web Flask

Abstract

Vehicle identification is a crucial component of traffic monitoring technology and transportation safety systems. Demand for solutions that can be detected in real time and accessed through web-based platforms has led to the emergence of a new approach based on efficient and adaptive deep learning. In this study, a vehicle detection system was designed using the YOLOv5 algorithm, known for its superior processing speed, smaller model size, and ease of implementation compared to previous versions of YOLO. The system was integrated with the Flask framework to provide a web interface that displays detection results from various video sources, both through live cameras and uploaded files. Based on testing, this system is able to recognize vehicles such as cars, buses, and trucks with a high level of accuracy and speed, while also presenting live vehicle count data. These findings prove that the combination of YOLOv5 and Flask can be an effective solution for implementing a web-based vehicle detection system with fast and accurate performance.